H‘ ‘I‘ 11 \ln 1 \ I ‘II M HIII» V M l .1 ‘2 W ‘E'ECHNEQUES FOR BANOENG WOODCOCK IN SUMMER HABETATS EN MiCHEGAN ‘z‘izesis far fire Oegme a? M & maxim 8W? ONNERSHY {3‘ y .2 7" H - . 2.353%"? :1. E33539 39?}? Y R A R B I L WW I V ,zjiuuawwmm mm: m w m a .. .9... Ala. m3 C w m .m U ABSTRACT TECHNIQUES FOR BANDING WOODCOCK IN SUMMER HABITATS IN MICHIGAN BY Larry E- Gregg Techniques for capturing American woodcock (Philohela minor) were evaluated during two summers in Michigan. A total of 457 woodcock were captured with mist nets, spotlights and ground traps. Mist nets were the most productive capture method tested, but night-lighting yielded a higher catch per unit of effort. Ground traps required a large investment in time and were productive only during dry weather. Woodcock were found to con- centrate in a variety of forest Openings on summer even- ings, including old fields, pulp cuttings and wood roads. Openings which attracted large numbers of woodcock were usually located near diurnal cover and contained areas of moist soil where the birds could probe for earthworms. Success in capturing woodcock depends upon their local population density, the availability of suitable banding sites, proper weather conditions and the experience of the banding crew. Recommendations for the establishment of a woodcock banding program are presented. TECHNIQUES FOR BANDING WOODCOCK IN SUMMER HABITATS IN MICHIGAN BY Larry E. Gregg A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Fisheries and Wildlife 1972 Xi H TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . iii LIST OF FIGURES. . . . . . . . . . . . . iv INTRODUCTION. . . . . . . . . . . . . . 1 STUDY AREAS O O O O O O O O O O O O O O 4 METHODS O 0 O O O O O 0 O O O O O O O 7 Mist Nets . . . . . . . . . . . . . 8 Funnel Traps . . . . . . . . . . . . lO Spotlights O O O O O O O 0 O O O O O 12 RESULTS 0 O O O 0 O O O O O O O O O I 14 Mist Nets . . . . . . . . . . . . . l4 Spotlighting . . . . . . . . . . . . 14 Ground Traps . . . . . . . . . . . . 17 Factors Influencing Capture Success. . . . . 19 DISCUSSION 0 O O O O O O O O O O O O O 30 LITERATURE CITED . . . . . . . . . . . . 36 ii LIST OF TABLES -Page Summer catches of woodcock, Ishpeming and Pigeon River areas, Michigan. . . . . . . 15 Relative efficiency of woodcock summer capture techniques, Ishpeming and Pigeon River areas, Michigan, 1965-66 . . . . . . . . . . 18 Comparison of woodcock singing-ground counts near Michigan summer study areas . . . . . 20 Age and sex ratios of woodcock captured by different techniques in Ishpeming and Pigeon River areas, Michigan, 1965-66 . . . . . . 22 Monthly weather conditions and capture success in Ishpeming and Pigeon River areas, Michigan . 26 iii LIST OF FIGURES Figure Page 1. Woodcock banding study areas in northern Michigan, 1965-66 . . . . . . . . . . 5 2. Net poles held in place by slipping them over short lengths of galvanized pipe driven into the ground . . . . . . . . . . . . 9 3. Funnel traps constructed of welded wire with cloth netting tops. . . . . . . . . . ll 4. Night—lighting equipment used to capture woodcock in Michigan . . . . . . . . . 13 5. Mist nets placed between tree clumps in wood- cock flight lanes generally produced good catches . . . . . . . . . . . . . 29 iv INTRODUCTION The popularity of the woodcock as a game bird is rapidly increasing, as evidenced by the change in conti— nental woodcock kill. Goudy (1967) estimated that more than a million woodcock were harvested in North America during the 1965 hunting season. The estimated total wood— cock kill has more than doubled during the past decade (Goudy and Martin, 1966). In Michigan, woodcock hunters now approximate waterfowl hunters in numbers. During the 1966 hunting season, more than 180,000 woodcock were harvested there by an estimated 85,750 hunters (Hawn, 1967). In spite of the increasing popularity of woodcock hunting, however, relatively little research or management effort is directed at the species. Though the utiliza- tion of banding data is an important element in the analy- sis of avian populations, woodcock bandings have been insufficient to provide the information needed to insure proper management. A summary of woodcock bandings and recoveries provided by the Bureau of Sport Fisheries and Wildlife's Migratory Bird Populations Station revealed that less than 20,000 woodcock had been banded up to 1965 and fewer than 400 bands had been recovered. In past years woodcock banding activities have centered on the birds' wintering grounds in Louisiana, where large numbers of birds occupy Open fields at night (Glasgow, 1958). Lights and hand nets have been employed in banding more than 17,000 woodcock on the Louisiana wintering grounds from 1948 to 1968 (Martin et_al,, 1969). Recent work within the birds' breeding range has revealed that woodcock also fly to clearing at dusk during the summer. Sheldon (1961) first reported woodcock concen— trating in fields at twilight in central Massachusetts. Similar observations were subsequently made by Martin (1962) at the Moosehorn National Wildlife Refuge in northwestern Maine and in the Upper Peninsula of Michigan (Martin and Clark, 1964). Sites used by woodcock on summer evenings have been described by Sheldon (1960, 1961) and Clark (1966). They found that fields used by woodcock varied consider— ably in size and vegetation. Sheldon (1960) stated that the only feature in which the fields were similar was their location with reference to spring breeding areas, all being in or close to places where the largest number of singing males was heard in the spring. The present two—summer study was initiated in Michigan in June, 1965, to determine the comparative value of several capture techniques and to provide an estimate of the feasibility of initiating a large-scale banding program in the state. The project was a COOperative one between the Department of Fisheries and Wildlife, Michigan State University, and the Bureau of Sport Fisheries and Wildlife, U. S. Department of the Interior. Assistance and use of facilities were provided by the Michigan Department of Natural Resources. I wish to acknowledge the advice and guidance pro- vided throughout the study by W. H. Goudy, U. S. Fish and Wildlife Service. Appreciation is also expressed to Dr. G. A. Petrides, Michigan State University, and to Dr. G. A. Ammann and T. R. Prawdzik, Michigan Department of Natural Resources, for their assistance. R. A. Hunt and J. B. Hale, Wisconsin Department of Natural Resources, also reviewed the manuscript. STUDY AREAS Field studies were conducted during the summers of 1965 and 1966, respectively, near Ishpeming in the central portion of the Upper Peninsula of Michigan and in the -Pigeon River State Forest in northern lower Michigan (Figure 1). Both study areas were characterized by a gently rolling topography. Sandy or sandy loam soils occurred on the well-drained sites with muck and peat in the lowlands. The original forests in both areas had been modified by logging, fire and agriculture. Vegetative cover was a mosaic of hardwood timber, grassy plains, coniferous swamps and plantations, hardwood-conifer mix- tures and old fields. Most of the land that was once farmed now lies fallow. Counts of singing male woodcock were made on routes followed annually through these areas. These tallies indicated that both supported sizable wood- cock breeding populations (Martin, 1963, 1964b; Goudy and Martin, 1966). Both the Pigeon River (Blankenship, 1957) and Ishpeming areas (Martin and Clark, 1964) had been studied during previous woodcock investigations. Since some of the IShpeming area banding sites utilized in the presents study had been previously discovered, field activites (SNPEMING AREA PlGfiofl mm 6‘“ Figure 1.--Woodcock banding study areas in northern Michigan, 1965-66. were undertaken at a new location in 1966 in order to pro- vide additional information concerning the ability of an investigator to identify suitable banding sites. METHODS Mist nets, funnel traps and spotlights were eval- uated for effectiveness as capture methods. The effects of weather, season, habitat and woodcock sex and age were evaluated with respect to capture success. Recapture data were examined for information on woodcock movements and habitat preferences. Birds were sexed and aged at the time of banding. Sex was ascertained by the combined width of the outer three primaries and by bill length (Greeley, 1953; Blankenship, 1957). Age was determined by flight feather molt and the color pattern on the tips of the secondaries (Martin, 1964a). As Sheldon (1961) suggested, sex and age ratios of captured woodcock may not be representation of those in the population. The reliability of sex and age ratio statistics derived from captured samples of woodcock were further evaluated in the present study. Temperature, precipitation, cloud cover and wind strength were recorded in the field. In addition, weather records were obtained from the U. S. Department of Com- merce weather stations in Marquette in 1965 and from the Michigan Department of Natural Resources fisheries research station near Vanderbilt in 1966. Data on field size, vegetation, soil and distance from diurnal cover were recorded for openings used by woodcock in Michigan. Estimates of availability of wood- cock food.and evidences of feeding in the openings were also recorded. Mist Nets The technique of using nets for capturing birds is centuries old. Dewar (1928), concerning the European woodcock (Scolopax rusticola), advises that."the poacher takes advantage of the fact that they leave their cover every evening by the same opening to set nets to catch them as they fly to their feeding ground." Sheldon (1961), upon discovering that the American woodcock also flew to open fields on summer evenings, used mist nets to capture them. In the present study, mist nets (Type C, lZ-meter, 4-tier, nylon) were suspended from 10-foot poles made of 3/4-inch metal conduit. The hollow poles were held in place by slipping them over 4-foot pieces of 3/8-inch galvanized pipe driven into the ground (Figure 2). Nets were placed in abandoned farm fields, along logging trails and back roads, across streams and on bridges. Netting was also attempted at several muddy sites which exhibited woodcock droppings and bill probe- marks. Although up to 35 nets were operated with one assistant on some evenings, generally ten to twenty nets ‘-. $3.; ‘ ‘ ,‘ . ‘ -. . . , - . 7 'f',t3{ 3:“, ".}:'. 5’“. $1;- . ., ‘- .' . 31.1" 'v ‘1', . I ‘(E’fill’ ‘ i‘f . I 1? .,v.:.;:“_'.1',-".’- .. « _-_ ..‘$“-:“-‘;%; iv“ ‘01‘ 5&5 " ‘ -"‘ H". :, ‘ ‘ ‘3", 1* '§ ‘ H 33—, .- :1 .b‘;;‘-g _‘ - . ~'-.' '1‘?“ )‘M—';' , \“ 1":~:‘.$-- I, 1 t; w $336 a1 ‘-"* gm... 3...». - ‘t mg: r. ~- -4..:M Figure 2.--Net pOles held in place by slipping them over short lengths of galvanized pipe driven into the ground. 10 were in Operation on an evening. .Nets were Opened shortly before sunset and furled at the cessation Of woodcock activity. In Massachusetts, a steady decline was noted (Sheldon, 1961) in the number Of birds using the fields after netting activities were initiated. To avoid inhib- iting woodcock use, in the present study nets were placed at more than one site and activities were rotated among sites. Funnel Traps Woodcock were captured in diurnal habitats by the use of small, wire ground traps similar tO those described by Liscinsky and Bailey (1955) and modified by Martin and Clark (1964). The traps were formed of 8-foot by 1—foot pieces Of 1—inch by 2-inch welded wire. Traps were usually formed into a kidney-shaped design, with a 6-inch entrance funnel Of welded wire (Figure 3). Trap tops were of cloth netting held in place by bending over strands of wire from the tOp edge of the trap. Fences Of poultry wire 1-foot high and up to 40-feet in length were used to direct walking birds into the traps. Strands of wire at the inner end of the entrance were bent toward the center to Oppose the escape of trapped birds. One to ten traps were placed at each Of several sites, most being in the predominantly alder (Alnus sp.) 11 o .11.-_‘._s__. ._. Figure 3.—-Funnel traps constructed of welded wire with cloth netting tOps. 12 cover type described by Mendall and Aldous (1943) as a favorite summer habitat for the bird. Spotlights Head-mounted spotlights powered by 6-volt dry—cell batteries were used to capture birds in fields on dark evenings. Birds often could be temporarily blinded after they flushed, causing them to alight promptly or to circle the light and finally to land nearby. The birds were then captured by means Of large, hand-held, landing nets. .Nets with 12-foot handles were used early in the study, but were later replaced with more maneuverable nets having 3-foot handles. .Netting and nets were dyed and painted black in an attempt to make them less conspicuous and to prevent re-flushing of woodcock "shined" to the ground (Figure 4). 13 Figure 4.--Night-1ighting equipment used tO capture woodcock in Michigan. o". a. RESULTS A total of 457 woodcock was captured during the study. Of these, 401 were banded and released, 46 were repeat captures or recaptures Of birds banded elsewhere, and 10 were casualties (Table 1). Mist Nets Mist nets caught more than 70 percent of the total (Table 2). At least one capture was made on 107 Of the 149 evenings that nets were Opened during the two summers. Woodcock were captured with nets at a variety of sites and under various weather conditions. Nights with rain showers or high winds were the only occasions when nets were not Operable. Most birds were captured during a brief period at dusk each evening, flight time normally lasting only five to fifteen minutes. The brief duration of daily netting Operations resulted in few birds of other species being caught. Spotlighting Although most birds were caught by mist nets, the best catch per man hour was accomplished using Spotlights. Woodcock were captured with lights, however, on only 20 nights during the two summers. Use of the method was 14 15 nmv o mm mwa v mm mmm mamuoa ms m ma «4 o m ma mange am 0 H m N 4 mm munmflauomm Hmm v m vHH N ma mma mpO27pmHz mpuflm mpuflm mousummo mOHUHOSmmo monsummomm Bmz mwflpawsmwo mwusummoom 3oz pozumz Hmuoa r1- musumwu coma mama Li I iL r}? ’i’ L yr V II ) I. V i? I .OOOAQOHS «mmmnmyum>flmvsommflm paw OOHEOQQmH ~03030003 mo mmnoumo HOEESm11.H mquB l6 limited by the need for near-total darkness. Birds were rarely seen on the ground before they flushed and many of those that flew were able to escape the light beams. Possibly, too, some birds did not flush and thus avoided capture. Martin (1962) reported that a bird dog flushed several woodcock in the field previously checked by a night-lighting crew. He estimated that only about one- third of the birds in the fields could be captured. In the present study, the proportion of birds flushed that were captured varied considerably between evenings, depending mainly upon the weather and upon characteristics of the capture site. If the evening was clear or windy, the technique was ineffective. Night—lighting was also generally unsuccessful in recently lumbered areas because ground cover was sufficient to allow downed birds to hide. It was Often possible to flush and to down a woodcock several times in such areas and yet not to succeed in capturing the bird. A 12-volt motorcycle battery and sealed beam spotlight used on one occasion during 1966 did not appreciably improve success, but the evening was less than ideal for the test. Success in night-lighting woodcock in West Virginia improved considerably when dry—cell batteries were abandoned in favor Of motorcycle batter- ies (Reiffenberger and Kletzly, 1967). 17 Ground Traps Funnel-type traps caught woodcock only infrequently during the study. Captures were recorded on only 38 per- cent Of the 149 days that traps were in Operation. Catch per man-hour with traps was the lowest of the three techniques used (Table 2). Trapping efficiency improved markedly during the second summer, but the catch rate still remained below that achieved with the other methods. In 1965, traps were placed randomly in what appeared tO be suitable habitat. In 1966, prospective trapping sites were examined, Often with the aid Of a pointing dog. Traps were then placed on or near the exact spots where birds were flushed. Captures Often occurred a day or so after trap erection when traps were distributed in the latter fashion. Traps were moved more frequently in 1966, too, since traps left in the same spot for long periods during the first summer yielded few captures. In addition to woodcock, 35 ruffed grouse (Bonasa umbellus) and four wood ducks (Aix sponsa) were captured in funnel traps. Thrushes and other ground-frequenting birds were also caught regularly. Red squrrels (Tamiasciurus hudsonicus) proved to be slight nuisances by chewing holes in the netting Of trap tops. -Rredation on trapped birds was also a problem, with one woodcock, five grouse and several passerine casualties recorded as 18 m.o ekma mm and smv moonpmz Haa «.0 one mm mqa msmwnmmuu SORH me wanna o.o mm mm em musonucma on Hm murmuapomm v.0 mom on HVH mnnmflalum: Naom Hmm mumz umflz usomwmmz pHOmmm mo Hawwmmowwm DHWWMM uHOmwm Omudummu Oonwmz xooopooz musomlcmz mmmucmoumm mama Ouzpmmu xooopooz Ousummu L FL) ['?7)' I I rDVLV-rlru'irtirlr 1r} rr F? r '" .mwtmmma .cmmHQOHz .mmwuw uw>wm commflm paw mcHEOmamH .mmsvflcnomu OHSDQOO HOaESm MOOOOOOS mo >OOOHOMMMO O>flpwammll.m mqmde 19 killed during the study. Known predators included a raccoon (Procyon lotor) and a mink (Mustela vison). Factors Influencing Capture Success Capture success for 174 days during two summers varied between zero and 15 woodcock per day. Daily capture success at a particular banding site depended upon: (1) the size and structure of the woodcock popula- tion in the area, (2) the attractiveness of the banding site to woodcock, (3) weather factors, (4) duration of capture effort at the site and (5) distribution Of capture devices. Numerical values could not be assigned to some of these variables, but a subjective analysis of the rela- tionships of these factors to capture success was attempted. ~Population densities.-—Woodcock population densi— ties On the two study areas were not known. Route count indexes to population size in these areas (Martin, 1963 and 1964b), though, averaged higher in Marquette County than in Cheboygan and Otsego Counties (Table 3) and prob— ably were representative Of a higher breeding population in the Upper Peninsula location. Average numbers Of captures per day in the Ishpeming and Pigeon River areas were approximately three birds and two birds, respec- tively. More than ten captures in one day were attained three times in the Ishpeming area but not once in the Pigeon River location. 20 TABLE 3.--Comparison of woodcock singing-ground counts near Michigan summer study areas.a Mean Birds-Per Stop Study Area Route County NO' 1963 1964 1965 Pigeon River 32 Otsego 0.57 0.71 0.71 73 Otsego 1.00 1.10 1.27 81 Cheboygan 0.57 0.71 o.c.b 83 Cheboygan 2.22 2.78 O.c. 98 Otsego 1.00 0.86 0.86 103 Cheboygan 2.27 2.00 O.c. Ishpeming 126 Marquette 2.00 2.25 1.86 131 Marquette - 2.57 2.14 aSources: Martin (1964b), Goudy and Martin (1966). b years. Observer change--count not comparable to previous 21 While woodcock numbers were evidently lower at Pigeon River than in the Ishpeming area, in all likeli- hood both areas supported populations that were better— than-average for Michigan as a whole. More than two birds per stOp were heard on the singing-ground transects located near the primary banding sites (Routes 83 and 131, Table 3). This is considerably higher than the average of 1.3 birds per stop heard on all Michigan routes in the same years. Age and sex composition.—-Immature birds com- prised 72 percent of the catch (Table 4).. The age ratio of woodcock captured during the study was 2.5 immatures per adult or 4.3 immatures per adult female. This ratio probably did not accurately represent the age structure of the population, since a woodcock clutch normally con- sists of only four eggs and a single brood is produced annually (Mendall and Aldous, 1943). The age ratio of captured woodcock was much higher than the 1.6 immatures per adult female found among fall-shot Michigan woodcock (Martin et_al., 1965; Goudy and Martin, 1966) during recent years. The sex ratio Of immature birds captured during the study was 1.4 males per female. More immature males than females were captured with all techniques employed (Table 4). On several evenings during the two summers, netting catches consisted exclusively Of juvenile males. 22 v.H s.o m.m oma msa as as mamnoe m.H m.o m.m SH mm ma 4 manna m.H m.H H.H AM «H OH ma mpnmwflpomm m.H s.o o.m ma Hma «4 mm mhmz-nmflz OHOEOm\OHm2 mamEOM\OHOE .O¢\.EEH OHOEOM OHOZ OHOEOM OHOE Begum: OHumm xmm‘ Oprm xmm.. OHOOM mod OHSDOEEH . padpm. Omsummu OHSDOEEH pascfi .molmomH .cmmHSOHz .mwmud HO>Hm commflm pom mcwfimmcmH :H mOSWHOSOOu #OOHOMMHO an Uwuspmmo Mooopooz mo mOHumH xmm UGO mm¢ll.v mqmme 23 Sheldon (1961) and Martin (1962) also noted a similar imbalance. If it may be presumed that the sex ratio at hatching is even and that immature males and females are taken in equal numbers by hunters cooperating in the Bureau's wing survey (Martin et_al., 1965), then either there is greater activity by young males during the summer or they show a greater preference than females for visiting forest Openings at that season- In contrast to immatures, the sex ratio among adults was 0.7 males per female. A similar adult sex ratio (0.64 males per female) has also been reported by Martin et_al. (1965) from fall—shot samples. Martin et_al, (1965) suggested that the preponderence of females might be due to higher losses among males as a result either of harsh weather during spring migration (when males migrate northward before females) or of predation during spring courtship displays. 'Physical factors.--Evening concentration sites differed markedly in physical characteristics and in their capacity to attract woodcock. Banding sites used by large numbers of woodcock shared the following features: (1) Openness, (2) Proximity to diurnal cover and (3) food availability. Birds were found in both lowland and upland forest Openings and these varied in size from considerably less than one acre to more than 30 acres. 24 Concentration sites were located close to diurnal woodcock cover, primarily alder, and were usually asso- ciated with a stream. All Of the evening fields found in this study were less than two miles distant from a stream and most were located one-half mile or less from extensive amounts of alder. Banding sites were not sampled for woodcock foods. However, birds were occasionally seen feeding and large numbers of probe marks in the mud at some sites gave further evidence that feeding did occur. Woodcock were attracted to areas where the soil had been recently dis- turbed, apparently visiting these sites to Obtain earth- worms. More than 25 birds were caught during a six-week period at a small muddy site created by a bulldozer during replacement of a culvert. Earth-moving activities involved with the construction of an earthen dam also attracted woodcock. Many woodcock were captured along tote roads within pulpwood harvest areas and along lightly travelled forest roads, especially following a rainfall. Woodcock were also found in dry upland sites at night, however, where it would have been nearly impossible for them to obtain earthworms. Sheldon (1961) analyzed stomach contents of woodcock captured in fields and found that a variety of fly and beetle larvae had been consumed by the birds. Sheldon reasoned that woodcock came to fields to Obtain animal food other than earthworms. 25 Krohn (1970) collected 60 woodcock in several dry, infer- tile Maine summer fields, however, and found little evi- dence of feeding by the birds. He concluded that wood— cock did not use the Maine fields primarily for feeding. Weather conditions almost certainly influenced capture success, but it was Often difficult to dis- tinguish their effects from that of the other variables. Weather during the summer Of 1965 was generally cool and rainfall was above average (Table 5). The summer of 1966 was characterized by average temperatures and well below normal precipitation. Strommen (1966) reported that "drought conditions continued to intensity over most of the State. Particularly hard hit was the northwest lower peninsula area where extreme drought, as expressed by the Palmer Index, was reached the last week Of July." Rain- fall was probably the most critical weather factor affect— ing capture success, since the presence Of moist soil for probing appeared prerequisite to woodcock use in some concentration sites. Light precipitation also seemed to stimulate more birds to engage in crepuscular flights, and misty evenings generally produced good catches. Lack of rainfall during the summer of 1966 seemed to be pri- marily responsible for the reduced netting and night- lighting success experienced that year. Temperatures at flight time ranged from 41° to 81°F. Although woodcock engaged in crepuscular flights 26 .Hoomav coaumum SOHmmmOH mmflumnmflm HO>Hm commflm paw Amomav Owumsvumz OH OOHumum Hmsgmmz monoEEOU mo wamEOHmmmo Eonm camp Hmnummzw mmmp o m H mH RH m m N m o -mmup OOH\m6HHm mm m Hm mm H OH m m H o mmusummo mmmoosw mafimmmue muson H m m m o h o m s m lama OH\muuHm o m m H o me o mH mH m mmusummo mmmoosm mcHuanHuamHz munmwa mH «H OH NH mm HH km mH MH s Ipmc OOH\m6HHm 50H mm mm mm mH How om em as HH mmusummo mmmoosm mcflgumz Hmauoq sv.m- Ho.H- om.o- Hm.H- mo.Hu mo.mn km.o+.Ho.ou OH.H- mm.H- scum mwusuumdmo ms.o mm.m mo.m mm.H mo.H mm.OH. mm.e mo.m OH.~ Hm.H mmnocH Hmuoe HOOHumuHmHOOHA. HOEHOG H.~- m.oa m.m+ H.H+. 0.4: ~.ma o.m- one scum musuummma o.Hm m.mo m.mw m.Ho H.mm w.mm H.mw H om mmmnm>¢ *mOHSOOHOQEOB mHOuOB .pmom .054 >HSU mcso mamuoa. .pmmm .m54 mash OCSO HOEEsm HmEESm oomH momH .cmmHSOHE .mOOHO HO>Hm commflm Ocm mCHEOman OH mmmoosm OHSOQOO paw chHDHOOOO chummz mHnucozln.m mHm